Before you can size a battery bank, a solar array, or the wire between them, you have to know what the load at a remote site actually looks like over a day and across a year. That description is the load profile, and it is far richer than a single number of watts. A remote instrument shed rarely draws a flat, steady load; it sips a small continuous base draw most of the time and then jumps for a few seconds when a radio transmits, a solenoid fires, or heat trace switches on. Characterizing that shape is the first real design step, because the average of the profile drives how much energy you must store and generate, while the peak of the profile drives how much current your controller and conductors must survive.
Remote Site Load Profile in one line: A remote site load profile is a description of how much power a site draws and how that draw changes over time, breaking the total into a continuous base load and the intermittent spikes layered on top, plus any seasonal variation. It matters because average load drives battery and solar sizing while peak load drives controller and wire sizing, and a single flat watts figure hides both. Characterizing the load profile is the step that feeds an honest watt-hour budget.
The first thing to separate in a load profile is the continuous base draw from the intermittent load riding on top of it. The base draw is what the site consumes every second it is powered: the RTU or flow computer running, a transmitter loop energized, a modem idling, a small controller awake. This part is easy to underestimate because it is unglamorous and quiet, yet it runs twenty-four hours a day and usually dominates the daily energy total. Intermittent loads are the opposite in character. A radio transmit burst, a solenoid actuation on a control valve, a sampler pump cycling, or a chart recorder motor draws real current but only for a short slice of each hour, so its contribution to daily energy depends entirely on how often and how long it fires.
The bridge between an intermittent load's momentary current and its real energy cost is its duty cycle, the fraction of time it is actually on. A ten watt radio that transmits for three seconds a minute contributes far less daily energy than its ten watt nameplate suggests, because it is off ninety-five percent of the time. Getting the duty cycle right for each intermittent load is where a load profile earns its keep, because assuming everything runs continuously grossly oversizes the system, while ignoring the spikes entirely undersizes the wire and controller that have to carry them. The honest approach is to list every load, note whether it is continuous or intermittent, and for the intermittent ones estimate both the current draw and the fraction of each hour it is active.
Seasonal variation is the component that catches people who characterize a load on a mild afternoon and never revisit it. Heater loads are the classic offender. Heat trace on an impulse line, a heated enclosure, an instrument heater, or a heated sample line may draw nothing for eight months and then become the single largest load on the site through winter, cycling on a thermostat as the temperature drops. A load profile that captures only the summer condition will size a battery and array that collapse in January, exactly when solar input is also at its weakest. A useful profile therefore describes at least a worst-case winter day and a typical day, so the design covers the season that actually stresses it.
The reason a flat watts number is dangerous is that two completely different design questions pull on two completely different features of the load profile. The average load, meaning the total energy over a day divided by the hours in that day, is what drives energy storage and generation. Your battery bank has to carry the site through the dark and cloudy stretches, and your solar array has to replace that energy plus the daily draw, so both are sized from the daily watt-hour total that the average load represents. If the average is wrong, the battery runs flat during a bad-weather run or the array can never catch up, and no amount of correct wiring saves it.
The peak load, meaning the largest instantaneous current the site can pull, drives a different set of components entirely. When the radio keys up at the same instant the solenoid fires and the heater is on, the conductors, the charge controller, the fuses, and any DC-DC converter all have to carry that combined surge without overheating or tripping, even though it lasts only seconds and barely moves the daily energy total. Sizing those components from the average would leave them undersized for the peak, so the wire runs hot and the controller current-limits or faults exactly when the site needs its radio. Peak load is about survival of a momentary event; average load is about surviving a week of weather.
Because these two numbers can be wildly different at a remote site, collapsing them into one figure of watts loses the information the design actually needs. A site might average fifteen watts yet peak at eighty watts for two seconds a minute. Report only the fifteen and someone buys thin wire that overheats on every transmit; report only the eighty and someone buys a battery bank four times larger than necessary. The load profile keeps both numbers alive and attached to their duty cycles, which is why it is the characterization step that everything downstream depends on. Get the profile right and the watt-hour budget, the battery bank, the array, and the conductor sizing all follow from it honestly.
A load profile can be estimated on paper from device nameplates and duty cycles, and that estimate is the right way to start a design. But the estimate is only as good as the assumptions behind it, and remote sites have a habit of drawing more than expected once heaters, comms retries, and forgotten auxiliary loads are all running together. The most reliable profile comes from measurement, either by logging the DC current at the battery terminals over a full day and, ideally, across a season, or by reading the load data the site itself already reports. A shunt or a battery monitor that records amp-hours in and out turns guesswork into a real curve you can size against.
This is where a monitoring platform earns its place in the power design rather than just the process design. A cloud SCADA system such as Merobix already trends battery voltage, charge current, and load current from many remote sites, which means the real load profile of a site is being recorded continuously rather than sampled once during commissioning. Trending that data reveals the things a paper estimate misses: the heater that cycles more aggressively than the datasheet implied, the comms module that retries and draws extra current when signal is poor, the slow creep in base load as instruments are added over the years. A profile drawn from months of actual site data is far more trustworthy than one built from nameplates alone.
Keeping the load profile visible after commissioning also protects the system as the site changes. Remote sites accumulate load. Someone adds a second transmitter, swaps a radio for a hungrier one, or installs a heater that was not in the original scope, and the once-comfortable battery bank starts going flat in bad weather with no obvious cause. When the load current is trended in SCADA, that creep shows up as a rising baseline long before the site fails, and the operator can respond by expanding the array or the bank rather than by dispatching a truck to a dead site. The load profile, in other words, is not a one-time calculation but a living characterization worth watching for the life of the site.
A flat watts number hides the two things the design actually needs. Average load, the daily energy total, drives battery and solar sizing, while peak load, the largest momentary current, drives wire, fuse, and controller sizing, and at a remote site those two can differ by five times or more. A site averaging fifteen watts might peak at eighty for a few seconds when the radio, solenoid, and heater coincide, so reporting only one figure oversizes or undersizes part of the system.
A duty cycle is the fraction of time an intermittent load is actually on. It matters because it converts a load's momentary current into real daily energy: a radio that draws ten watts but transmits three seconds a minute contributes far less energy than its nameplate suggests. Getting each intermittent load's duty cycle right is what keeps a load profile from wildly oversizing the battery and array by pretending every device runs continuously.
Heaters introduce strong seasonal variation. Heat trace, heated enclosures, and heated sample lines may draw nothing for most of the year and then become the largest single load through winter, cycling on a thermostat as temperature drops. Because winter is also when solar input is weakest, a profile built only from a mild-weather measurement will size a system that fails in January, so a good profile always includes a worst-case winter day.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.